A while ago, I posted examples of NR in a portrait of the lovely Silla, Fairy
of the Burren of Ireland.
I've since done a fair amount of informal testing of NR on various subjects. Also known as "Will this work? Nah. How
about this? Hmmm. OK."
Then, I actually looked into what Topaz has had to say about their two pronged attack on noise. In Topaz Denoise AI,
there are two processing options, Denoise and Clear. I hadn't paid Clear much attention, thinking I was familiar with it
from earlier versions.
Well, I was wrong. Denoise was trained on lower noise examples, Clear on more noisy ones. On the files I mostly work on,
GX9, ZS200 and A7, ISO 800, or lower ISOs with pulled shadows is about the max for Denoise. Above that, Clear AI comes
into its own.
My sample is ISO 3200, on a 2/3 size sensor. So, I've added a sample done with Clear AI.
<http://www.moosemystic.net/Gallery/tech/Process/NR/Some%20NR.htm>
Hmmm, right up there with the DxO PL2. I could argue either way, I think. DxO's tungsten WB was awful, very yellow; I
couldn't "see through" it, so I worked on it. The ACR WB underlying the other examples was decent to start with, so I
left it alone. Until now, when two are so close for NR, so color will affect opinions, so I've also adjusted the WB for
the sample using Clear AI.
Off Color Moose
--
What if the Hokey Pokey *IS* what it's all about?
--
_________________________________________________________________
Options: http://lists.thomasclausen.net/mailman/listinfo/olympus
Archives: http://lists.thomasclausen.net/mailman/private/olympus/
Themed Olympus Photo Exhibition: http://www.tope.nl/
|